Current implementation looks like this:

public byte[] getBytes(String charsetName)
        throws UnsupportedEncodingException {
    if (charsetName == null) throw new NullPointerException();
    return encode(lookupCharset(charsetName), coder(), value);
}

Null check seems to be redundant here because the same check fo 'charsetName' is done within String.lookupCharset(String):

private static Charset lookupCharset(String csn) throws UnsupportedEncodingException {
    Objects.requireNonNull(csn);
    try {
        return Charset.forName(csn);
    } catch (UnsupportedCharsetException | IllegalCharsetNameException x) {
        throw new UnsupportedEncodingException(csn);
    }
}